Energy & Fuels 2008, 22, 3571–3574
3571
Surface Properties of TiO2/Al2O3 To Synthesize Methyl Propyl Carbonate via Gas-Phase Transesterification Xingquan Chen,*,†,‡ Yanming Dong,† Chunxiang Zhao,‡ and Tiansheng Zhao‡ Jiangsu Key Laboratory of Finechemical Technology, Jiangsu Polytechnic UniVersity, Jiangsu, Changzhou 213016, China, and Key Laboratory of Energy Resources and Chemical Engineering of Ningxia UniVersity, Ningxia, Yinchuan 750021, China ReceiVed April 24, 2008. ReVised Manuscript ReceiVed September 9, 2008
Methyl propyl carbonate (MPC) was produced by transesterification of DMC and propanol in a continuous gas flow reactor over heterogeneous TiO2 catalyst supported on Al2O3. Among different Ti loadings of TiO2/ Al2O3 catalysts, 5 wt % Ti loadings of TiO2/Al2O3 showed the highest activity. The morphological and chemical states of titanium species in TiO2/Al2O3 were characterized by means of X-ray diffraction (XRD), Brunauer-Emmett-Teller (BET), temperature programmed desorption (TPD), and Fourier transform infrared spectroscopy (FTIR), respectively. The catalyst with 5% Ti loadings showed high stability within 130 h, and MPC yields and DMC conversion were 63 and 51%, respectively.
1. Introduction Methyl propyl carbonate (MPC) is an asymmetrical methyl alkyl carbonate, which is a promising single solvent for Li ion battery electrolytes.1 In addition, it is used for organic synthesis because it has a methy group, propyl group, and carbonyl group in its molecule. At present, the esterification of methyl chloroformate with alcohol is the main synthesis method for methyl alkyl carbonate.2,3 However, the esterification route is a not environmentally benign process because the highly toxic methyl chloroformate is used and the process brings about co-production of HCl. Dimethyl carbonate (DMC) is a nontoxic building block that can be used in organic syntheses as a green substitute for toxic and corrosive reagents, such as dimethyl sulfate, methyl chloroformate, and phosgene.4-6 Therefore, DMC may act as a methylating agent, reacting with dipropyl carbonate or propanol to form MPC. In the transesterification of DMC with dipropyl carbonate, the reaction system is not a economic route because dipropyl carbonate is quite expensive. In addition, it is a critical thermodynamic limitation. Therefore, a new approach of transesterification of DMC and propanol became more attractive to us. In general, transesterification reactions are carried out in the liquid phase using homogeneous catalysts, such as acids and bases, organic Pb, Sn, or Ti compounds, and so on.7-9 However, there is an obvious disadvantage because of the difficulty of catalyst-product separation. Therefore, the development of active solid catalysts is highly desirable. As a few reports on the development of active solid catalysts indicate, TiO2/SiO2 * To whom correspondence should be addressed. E-mail: chenxqnxu@ 163.com. † Jiangsu Polytechnic University. ‡ Key Laboratory of Energy Resources and Chemical Engineering of Ningxia University. (1) Ein-Eli, Y.; Thomas, S. R.; Chadha, R.; Blakley, T. J.; Koch, V. R. J. Electrochem. Soc. 1997, 144 (7), 180. (2) Edmund, P. W.; Frank, C. W. C. J. Org. Chem. 1986, 51 (19), 3704. (3) Ichiro, M.; Jiro, T. Tetrahedron 1987, 43 (17), 3903. (4) Ono, Y. Catal. Today 1997, 35 (1-2), 25. (5) Ono, Y. Appl. Catal., A 1997, 155 (2), 133. (6) Gong, J. L.; Ma, X. B.; Wang, S. P. Appl. Catal., A 2007, 316 (1), 1.
and MoO3/SiO2 are found to be effective in the synthesis of methyl phenyl carbonate and diphenyl carbonate from DMC and phenol.9-12 However, there are very few reports on the synthesis of MPC. In this work, the transesterification of DMC and propanol was performed in a continuous gas flow fixed-bed reactor over solid catalysts at high temperatures, which were found to be more favorable from the thermodynamic calculation of the reaction of the synthesis of MPC. Among various catalysts tested, 5 wt % Ti loadings of TiO2/Al2O3, which has been proven to be a better catalyst in the transesterification reaction,13 was the most active catalyst for the transesterification of DMC and propanol in the gas phase.14,15 We employed BrunauerEmmett-Teller (BET), X-ray diffraction (XRD), Fourier transform infrared spectroscopy (FTIR), and temperature programmed desorption (TPD) to investigate the structure of the active components and the correlation between the structure and activity of the TiO2/Al2O3 catalyst in the gas-phase transesterification of DMC and propanol. 2. Experimental Section 2.1. Catalysts Preparation. TiO2/Al2O3 catalysts were prepared by the impregnation method using Al2O3 as a support. Al2O3 and other chemical reagent used for catalyst preparation were commercial products with C.P. grade. (7) Niu, H. Y.; Guo, H. M.; Yao, J.; Wang, Y.; Wang, G. Y. J. Mol. Catal. A: Chem. 2006, 259 (1-2), 292. (8) Mei, F. M.; Li, G. X.; Nie, J.; Xu, H. B. J. Mol. Catal. A: Chem. 2002, 184 (1-2), 465. (9) Fu, Z. H.; One, Y. S. J. Mol. Catal. A: Chemical. 1997, 118 (3), 293. (10) Kim, W. B.; Lee, J. S. J. Catal. 1999, 185 (2), 307. (11) Kim, W. B.; Lee, J. S. Catal. Lett. 1999, 59 (1), 83. (12) Kim, W. B.; Kim, Y. G.; Lee, J. S. Appl. Catal., A 2000, 194–195, 403. (13) Lacome, T.; Hillion, G.; Delfort, B.; Revel, R.; Lepord, S.; Acakpo, G. U.S. Patent 2005/0266139A1, Dec 1, 2005. (14) Zhao, C. X.; Chen, X. Q.; Zhao, T. S. Petrochem. Technol. 2005, 34 (1), 14. (15) Chen, X. Q.; Zhao, C. X.; Zhao, T. S. Petrochem. Technol. 2007, 36 (5), 447.
10.1021/ef800280a CCC: $40.75 2008 American Chemical Society Published on Web 10/16/2008
3572 Energy & Fuels, Vol. 22, No. 6, 2008
Figure 1. Continuous flow fixed-bed reactor for the synthesis of MPC: (1) N2 cylinder, (2) precontrol pressure, (3) mass flow meter, (4) input material bunker, (5) syringe, (6) preheater, (7) reactor, (8) condenser, (9) separator, (10) postcontrol pressure, and (11) online GC.
Al2O3 particles were impregnated with a solution of tetrabutyl titanate dissolved in ethanol. Samples of metal loadings given in weight percent by metal were dried in an oven at 383 K for 6 h to remove the organic solvent and calcined in a muffle furnace at 773 K for 3 h. 2.2. Gas-Phase Transesterification of DMC and Propanol. The transesterification of DMC and propanol was carried out in a continuous flow system with a fixed bed reactor. The reactor was made of a stainless-steel tube, with a diameter of 12 mm and a length of 450 mm. The schematic diagram of the experimental system is shown in Figure 1. A 5.0 g catalyst sample (20-40 mesh) was packed in the middle of the reactor and placed between two layers of glass bead beds. The temperature of the catalyst bed in the reactor was measured by a thermocouple and controlled with a precision of (1 °C by a temperature controller. The reaction pressure was measured by a digital pressure indicator and maintained constant by a back- and prepressure regulator. The reactant mixture composed of DMC and propanol, with a ratio of 1 mol of DMC to 1 mol of propanol, was introduced using a syringe pump (SZB-2) to the preheater, where it was vaporized and then entered the reactor together with nitrogen. The flow rate of nitrogen was controlled by a mass flow controller (D07-12A/ZM). The products were analyzed using the gas chromatograph (SP2100) equipped with a flame ionization detector (FID). A PEG20M (3 mm × 2.5 m) column was used for the analysis of liquid products. 2.3. Characterization of the TiO2/Al2O3 Catalysts. The specific surface area was analyzed using the BET method (Micromeritics ASAP 2010 M surface area analyzer). The IR spectroscopic measurements of adsorbed pyridine were carried out on a Nicolet NEXUS-470 FTIR spectrometer. The scanning range was from 400 to 2000 cm-1, and the resolution was 4 cm-1. The sample powder was pressed into a self-supporting wafer. Prior to each experiment, the catalysts were evacuated (0.67 Pa) at 573 K for 4 h. Then, they were heated to 303 K for 2 h. After this, the material was exposed to 30 Torr of pyridine for 30 min and finally evacuated for an additional 1 h at 473 K. After adsorption, the samples were outgassed and the spectra were recorded at room temperature. NH3-TPD and CO2-TPD spectra were recorded using a TP5000 chemical adsorption spectrometer. The catalysts was heated to 423 K in flowing He for 1 h and then cooled to room temperature. The gases adsorption (NH3 or CO2) was carried out at room temperature to saturation. NH3 or CO2 was replaced with argon, and the sample was heated to 873 K at a rate of 10 K min-1. Powder XRD measurements were conducted on an X-ray diffractometer of model SA-HF3 of Rigaku using a radiation source of Cu KR radiation (λ ) 0.154 nm) at 40 kV and 30 mA, with a scanning rate of 10° min-1.
3. Results and Discussion 3.1. Performance of Catalysts. The transesterification of DMC was carried out at 403 K under 1.1 MPa pressure using TiO2/Al2O3. Figure 2 showed DMC conversion and DPC yield
Chen et al.
Figure 2. Catalytic activity of TiO2/Al2O3 with different Ti loadings. Reaction conditions: reactant mixture feeding rate, 5 mL/h; the mole ratio of DMC/propanol, 1; N2 flow rate, 30 mL/min; reaction temperature, 403 K; reaction pressure, 1.1 MPa.
Figure 3. X-ray diffraction patterns of TiO2/Al2O3 with different Ti loadings: (a) Al2O3, (b) 1%, (c) 3%, (d) 5%, (e) 10%, (f) 20%, and (g) TiO2.
over TiO2/Al2O3 catalysts, with Ti loadings in the range of 0-20 wt %. We reported, in the previous paper,14 that Al2O3 was a low active catalyst, giving 1.0% DMC conversion, for the transesterification of DMC with phenol. As shown in Figure 2, the conversion of DMC was improved from 1.0 to 54.3%, when the amount of Ti loadings was increased to 5 wt %. It was remarkable that the DPC yield still remained about 47%. At Ti higher than 5 wt %, both the conversion of DMC and the yield of DPC decreased. It was interesting to see how this structural information could be related to the reactivity of TiO2/Al2O3 in the gas-phase transesterification of DMC and propanol. 3.2. Crystal-Phase Specific and Surface Area. Figure 3 showed the XRD patterns of TiO2/Al2O3 catalysts with the different Ti loadings. At low Ti loadings, no titanium phase was observed and it was possible for the surface TiIV species to form the small particles or become highly dispersed on Al2O3, thus resulting in a decrement of the specific surface area of the doped catalysts. However, characteristic peaks at the Bragg angle 2θ of 25.3 ascribed to the crystal phase of TiO2 in anatase could be detected in XRD patterns when the Ti concentration was higher than 3 wt %, suggested that the amount of dopant exceeded the largest capacity of the single-layer distribution on the Al2O3 surface. Figure 4 illustrated change in specific surface area for TiO2/ Al2O3 catalysts with increasing Ti loadings. From Figure 4, we can see that the TiO2/Al2O3 catalysts showed an almost linear decrease of BET surface areas as the Ti content increased up to 20 wt %. It can affect the BET of catalysts when TiO2 deposited on the surface or penetrated into
TiO2/Al2O3 To Synthesize MPC
Figure 4. BET surface area of the TiO2/Al2O3 catalysts with different Ti loadings.
Figure 5. NH3-TPD patterns of TiO2/Al2O3 with different Ti loadings: (a) 1%, (b) 3%, (c) 5%, and (d) 10%.
the body phase of Al2O3 because of the mixing of alumina with low surface area TiO2. Figure 2 showed that DMC conversion and MPC yield both showed maxima at 5 wt % Ti loading. This appeared to be due to the decrease in surface area at the high loading as shown in Figure 4 and TiO2 in the anatase phase on the surface as shown in Figure 3. 3.3. TPD of Ammonia Analysis. TPD of probe molecules, such as ammonia or pyridine, is a well-known method for the determination of acidity of solid catalysts because it is easy and reproducible. Ammonia is used frequently as a probe molecule because of its small molecular size, stability, and strong basic strength (pKa ) 9.2). The total acidity measurements of TiO2/ Al2O3 samples had been carried out by TPD of NH3. The acid strength distribution was classified depending upon the desorption temperature of NH3. The desorption temperature region for physisorbed is 353-423 K; weak, 423-523 K; medium, 523-623 K; and strong, 623-723 K.16 The TPD results of various TiO2/Al2O3 catalysts were given in Figure 5. From the results shown in Figure 5, it could be seen that the peaks appeared only in the low-temperature region, confirming that there existed only weak acid sites on the surface of TiO2/ Al2O3 catalysts, with Ti loadings ranging from 1 to 10 wt %. We carried out a preliminary study on the effect of acid strength on the catalyst, and a part of the results was reported elsewhere.15 3.4. IR Characterization of Adsorbed Pyridine. FTIR analysis of adsorbed pyridine allows for a clear distinction between Bro¨nsted and Lewis acid sites. The IR absorption bands at 1545 and 1455 cm-1 are assigned to adsorb pyridine (16) Satsuma, A.; Kamiya, Y.; Westi, Y.; Hattori, T. Appl. Catal., A 2000, 194, 195–253.
Energy & Fuels, Vol. 22, No. 6, 2008 3573
Figure 6. Adsorbed pyridine IR spectra of TiO2/Al2O3 catalysts with different Ti loadings: (a) Al2O3, (b) 1 wt %, (c) 3 wt %, (d) 5 wt %, and (e) 10 wt %.
Figure 7. Relative acid amount of TiO2/Al2O3 catalysts with different Ti loadings.
coordinated with Bro¨nsted acid sites and Lewis acid sites, respectively. The peak at 1490 cm-1 can be ascribed to the overlapping of Bro¨nsted acid and Lewis acid sites.17,18 From Figure 6, it can be seen that IR pyridine adsorption spectra of TiO2/Al2O3 has peaks at 1452 and 1490 cm-1, while the peak at 1545 cm-1 is absent, which means that there are only Lewis acid sites but no Bro¨nsted acid sites on the catalyst. Adsorbance of residual pyridine reflects the intensity of Lewis acid sites after desorption at evacuation temperatures. Figure 7 showed the relative acid amount of TiO2/Al2O3 catalysts with different Ti loadings. On the basis of the results from XRD and FTIR measurements, for TiO2/Al2O3 catalysts with TiO2 content ranging from 1-3 wt %, the amount of the total weak acid sites decreased with the amorphous TiO2 amount. However, crystalline TiO2 could provide a few acid sites by itself, which resulted in the increase of total acid sites on TiO2/Al2O3 catalysts at the higher TiO2 content with the presence of crystal TiO2.19 Although Ma et al. thought that TiO2 content affects the acidity of the catalysts and Lewis acid sites, which played important roles in the transesterification of DMO with phenol to produce MPO and DPO over TiO2/Al2O3,20 we could not find that catalytic activity had a closed relation with the total weak acid sites in the transesterification of DMC with propanol from Figures 2 and 6. 3.5. TPD of Ammonia Analysis. CO2-TPD characterization was conducted to survey the base strength of catalyst. The peaks (17) Emeis, C. A. J. Catal. 1993, 141 (2), 347. (18) Thomas, R. H.; Harry, M. W. J. Phys. Chem. 1967, 71 (7), 2192. (19) Liu, Y.; Ma, X. B.; Wang, S. P.; Gong, J. L. Appl. Catal., B 2007, 77 (1-2), 125. (20) Ma, X. B.; Wang, S. P.; Gong, J. L.; Yang, X.; Xu, G. H. J. Mol. Catal. A: Chem. 2004, 222 (1-2), 183.
3574 Energy & Fuels, Vol. 22, No. 6, 2008
Figure 8. CO2-TPD patterns of TiO2/Al2O3 with different Ti loadings: (a) 1%, (b) 3%, (c) 5%, and (d) 10%.
in the high-temperature region above about 673 K can be attributed to the desorption of CO2 from strong Bro¨nsted and Lewis base sites; the peaks in the temperature range between 450 and 673 K can be attributed to the desorption of CO2 from intermediate base sites; and the peaks in the temperature region below 450 K can be assigned to the desorption of CO2 from weak base sites, respectively. Ma et al. thought that the peak positions of TiO2/Al2O3 were consistent with those of Al2O3, while the amount of basic sites on TiO2/Al2O3 was a little more than that on Al2O3.20 Figure 8 showed CO2-TPD patterns of TiO2/Al2O3 with different Ti loadings. From the results shown in Figure 8, the significant peaks appeared in the 150 and 210 °C temperature regions simultaneously. Therefore, there must exist both weak and intermediate base sites on catalysts, while the amount of weak base sites was more than that of intermediate base sites. It could also be observed that the characteristics of base sites varied slightly when different Ti loadings were supported on Al2O3, except that base sites decreased in 10 wt % TiO/Al2O3 might be due to the formation of the anatase phase. The conversions of DMC also decreased up to this loading and levels off at higher Ti loadings. This indicated that the base sites were responsible for activity during the transesterification reaction. 3.6. Stability of the TiO2/Al2O3 Catalyst. Under conditions [403 K; n (DMC)/n (propanol), 0.5; gas hourly space velocity (GHSV), 360 h-1], the stability of the TiO2/Al2O3 catalyst was examined as shown in Figure 9. During the 130 h stability test,
Chen et al.
Figure 9. Stability of TiO2/Al2O3 catalyst on transesterification. Reaction conditions: reactant mixture feeding rate, 5 mL/h; the mole ratio of DMC/propanol, 2; GHVS, 360 h-1; reaction temperature, 403 K; reaction pressure, 1.1 MPa.
DMC conversion slightly decreased from 64 to 62% within 50 h and remained constant at longer reaction times. The selectivity of MPC did not change with the time on stream. MPC was the most predominant product, and the main byproduct was dipropyl carbonate. The selectivity of MPC was no less than 83%. This result clearly demonstrated that TiO2/Al2O3 showed a stable activity and high MPC selectivity. 4. Conclusions In this work, MPC synthesis reaction by transesterification of DMC and propanol was studied using a vapor-phase flow reaction system over an alumina-supported titania catalyst. The NH3-TPD and FTIR characterization showed that there existed only Lewis weak acid sites on the surface of TiO2/Al2O3. The CO2-TPD characterization indicated that the base sites were responsible for activity during the transesterification reaction. The catalyst with 5% Ti loading showed high MPC yields, DMC conversion, and stability with 130 h. Acknowledgment. This research was financially supported by the basic research project of the Ministry of Science and Technology of the People’s Republic of China (2002-016-02) and the project of China Jiangsu Minisitry of Education (06KJB150023). EF800280A
